ABSTRACT
The rapid advance of nanotechnology has led to the development and widespread application of nanomaterials, raising concerns regarding their potential adverse effects on human health and the environment. Traditional (experimental) methods for assessing the nanoparticles (NPs) safety are time-consuming, expensive, and resource-intensive, and raise ethical concerns due to their reliance on animals. To address these challenges, we propose an in silico workflow that serves as an alternative or complementary approach to conventional hazard and risk assessment strategies, which incorporates state-of-the-art computational methodologies. In this study we present an automated machine learning (autoML) scheme that employs dose-response toxicity data for silver (Ag), titanium dioxide (TiO2), and copper oxide (CuO) NPs. This model is further enriched with atomistic descriptors to capture the NPs' underlying structural properties. To overcome the issue of limited data availability, synthetic data generation techniques are used. These techniques help in broadening the dataset, thus improving the representation of different NP classes. A key aspect of this approach is a novel three-step applicability domain method (which includes the development of a local similarity approach) that enhances user confidence in the results by evaluating the prediction's reliability. We anticipate that this approach will significantly expedite the nanosafety assessment process enabling regulation to keep pace with innovation, and will provide valuable insights for the design and development of safe and sustainable NPs. The ML model developed in this study is made available to the scientific community as an easy-to-use web-service through the Enalos Cloud Platform (www.enaloscloud.novamechanics.com/sabydoma/safenanoscope/), facilitating broader access and collaborative advancements in nanosafety.
ABSTRACT
ASCOT (an acronym derived from Ag-Silver, Copper Oxide, Titanium Oxide) is a user-friendly web tool for digital construction of electrically neutral, energy-minimized spherical nanoparticles (NPs) of Ag, CuO, and TiO2 (both Anatase and Rutile forms) in vacuum, integrated into the Enalos Cloud Platform (https://www.enaloscloud.novamechanics.com/sabydoma/ascot/). ASCOT calculates critical atomistic descriptors such as average potential energy per atom, average coordination number, common neighbour parameter (used for structural classification in simulations of crystalline phases), and hexatic order parameter (which measures how closely the local environment around a particle resembles perfect hexatic symmetry) for both core (over 4 Å from the surface) and shell (within 4 Å of the surface) regions of the NPs. These atomistic descriptors assist in predicting the most stable NP size based on lowest per atom energy and serve as inputs for developing machine learning models to predict the toxicity of these nanomaterials. ASCOT's automated backend requires minimal user input in order to construct the digital NPs: inputs needed are the material type (Ag, CuO, TiO2-Anatase, TiO2-Rutile), target diameter, a Force-Field from a pre-validated list, and the energy minimization parameters, with the tool providing a set of default values for novice users.
ABSTRACT
We present a new method for calculating the diffusion tensor for the systems of sorbates inside nanoporous materials at different loadings by just using transition rate constants. In addition, a user-friendly program with graphical user interface has been developed and is freely provided to be used (https://sourceforge.net/projects/kobra/). It needs from the user just to provide the values of the unit cell lengths and angles, the transition rate constants for each sorbate, and any spatial constraint between these sorbates. This program is shown to be about 30 times faster than kinetic Monte Carlo method. Application of the method to the problem of diffusion of aromatics in silicalite-1 at different loadings is presented too. © 2019 Wiley Periodicals, Inc.
ABSTRACT
We present a new method for solving the master equation for a system evolving on a spatially periodic network of states. The network contains 2(ν) images of a "unit cell" of n states, arranged along one direction with periodic boundary conditions at the ends. We analyze the structure of the symmetrized (2(ν)n) × (2(ν)n) rate constant matrix for this system and derive a recursive scheme for determining its eigenvalues and eigenvectors, and therefore analytically expressing the time-dependent probabilities of all states in the network, based on diagonalizations of n × n matrices formed by consideration of a single unit cell. We apply our new method to the problem of low-temperature, low-occupancy diffusion of xenon in the zeolite silicalite-1 using the states, interstate transitions, and transition state theory-based rate constants previously derived by June et al. [J. Phys. Chem. 95, 8866 (1991)]. The new method yields a diffusion tensor for this system which differs by less than 3% from the values derived previously via kinetic Monte Carlo (KMC) simulations and confirmed by new KMC simulations conducted in the present work. The computational requirements of the new method are compared against those of KMC, numerical solution of the master equation by the Euler method, and direct molecular dynamics. In the problem of diffusion of xenon in silicalite-1, the new method is shown to be faster than these alternative methods by factors of about 3.177 × 10(4), 4.237 × 10(3), and 1.75 × 10(7), respectively. The computational savings and ease of setting up calculations afforded by the new method of master equation solution by recursive reduction of dimensionality in diagonalizing the rate constant matrix make it attractive as a means of predicting long-time dynamical phenomena in spatially periodic systems from atomic-level information.
